NATIONALPARK-FORSCHUNG IN DER SCHWEIZ 89 (2000)

Schütz, M.; Krüsi, B.O.; Edwards, .PJ. (eds): Natl.park-Forsch. Schweiz 89, 207 - 235

Predicting the development of subalpine grassland in the Swiss National Park: how to build a succession model based on data from long-term permanent plots

Entwicklungsprognose für subalpines Grünland im Schweizerischen Nationalpark: der Weg von Dauerflächendaten zu einem Langzeitsukzessionsmodell

Martin Schütz¹, Otto Wildi¹, Gérald Achermann¹, Bertil O. Krüsi¹, Bernhard Nievergelt²

¹ Swiss Federal Research Institute WSL, CH-8903 Birmensdorf, Switzerland
² Institute of Zoology, Wildlife Research and Conservation Biology, University of Zurich, CH-8057 Zurich, Switzerland

Abstract

In the Swiss National Park, the development of the vegetation has been recorded on more than 160 permanent plots for 80 years. We assembled the data in 59 time-series after summarising the 256 descriptors (vascular plant species) into six species groups. The generated model covers a period of 585 years. This succession model predicts the development of the vegetation on the meadows and pastures in the subalpine vegetation belt of the Park. It describes the development of the vegetation from tall-herb communities through several stages of meadows and pastures to mountain pine forests. The direction of succession predicted by the model, i.e. the sequence of six successive plant communities, exactly matches the successional sequences observed on the individual permanent plots. The rate of succession on an individual plot, however, cannot be predicted accurately by the model: a certain plant communitiy may be stable and long-lived in one location while it is short-lived in another. If the descriptors are summarised less severely, e.g. to seven or more species groups instead of six, the direction of succession is no longer clearly defined, and the pathways of succession differ locally. On the whole, the long-term development of the vegetation in the Swiss National Park is best described by unimodal models. Linear or sigmoid models depict, at best, the short-term development of the vegetation correctly.

Martin Schütz1, Otto Wildi1, Gérald Achermann1, Bertil O. Krüsi1, Bernhard Nievergelt2

Abstract

Martin Schütz¹, Otto Wildi¹, Gérald Achermann¹, Bertil O. Krüsi¹, Bernhard Nievergelt²